Genetic correlation and causal associations between psychiatric disorders and lung cancer risk

Unsupervised discovery of clinical disease signatures using probabilistic independence

OBJECTIVE

This study uses probabilistic independence to disentangle patient-specific sources of disease and their signatures in Electronic Health Record (EHR) data.

MATERIALS AND METHODS

We model a disease source as an unobserved root node in the causal graph of observed EHR variables (laboratory test results, medication exposures, billing codes, and demographics), and a signature as the set of downstream effects that a given source has on those observed variables. We used probabilistic independence to infer 2000 sources and their signatures from 9195 variables in 630,000 cross-sectional training instances sampled at random times from 269,099 longitudinal patient records. We evaluated the learned sources by using them to infer and explain the causes of benign vs. malignant pulmonary nodules in 13,252 records, comparing the inferred causes to an external reference list and other medical literature. We compared models trained by three different algorithms and used corresponding models trained directly from the observed variables as baselines.

RESULTS

The model recovered 92% of malignant and 30% of benign causes in the reference standard. Of the top 20 inferred causes of malignancy, 14 were not listed in the reference standard, but had supporting evidence in the literature, as did 11 of the top 20 inferred causes of benign nodules. The model decomposed listed malignant causes by an average factor of 5.5 and benign causes by 4.1, with most stratifying by disease course or treatment regimen. Predictive accuracy of causal predictive models trained on source expressions (Random Forest AUC 0.788) was similar to (p = 0.058) their associational baselines (0.738).

DISCUSSION

Most of the unrecovered causes were due to the rarity of the condition or lack of sufficient detail in the input data. Surprisingly, the causal model found many patients with apparently undiagnosed cancer as the source of the malignant nodules. Causal model AUC also suggests that some sources remained undiscovered in this cohort.

CONCLUSION

These promising results demonstrate the potential of using probabilistic independence to disentangle complex clinical signatures from noisy, asynchronous, and incomplete EHR data that represent the confluence of multiple simultaneous conditions, and to identify patient-specific causes that support precise treatment decisions.

Letter to the FDA Proposing Major Changes in the US Clozapine Package Insert Supported by Clozapine Experts Worldwide. Part II: A Review of Fatal Outcomes in US Pharmacovigilance Data and Proposed Changes

PURPOSE/BACKGROUND

This is the second part of a 2-part article that proposes improving the United States (US) clozapine package insert. Part II focuses on fatal outcomes and the 5 boxed warnings, 4 specifically for clozapine: severe neutropenia, seizure, orthostatic hypotension and myocarditis, and 1 for all antipsychotics (elderly with dementia).

METHODS

US reports to the World Health Organization's global pharmacovigilance database were analyzed from clozapine's introduction to January 15, 2023.

FINDINGS/RESULTS

The US was the top reporter worldwide for clozapine with 56,003 reports and 9587 associated fatal outcomes. The 4 clozapine boxed warnings were associated with 534 fatal outcomes (218 with severe neutropenia, 131 with seizures, 125 with orthostasis, 36 with myocarditis, 24 with cardiomyopathy, and 0 with mitral valve prolapse). With no boxed warnings, pneumonia was associated with 674 fatal outcomes and increased white blood cell count (a sign of infection) with 596 fatal outcomes. After considering overlaps, pneumonia and increases in white blood cell count explained 900 fatalities, or 9.4% of 9587 fatal outcomes. The Food and Drug Administration continues to focus on severe neutropenia which was associated with only 218 or 2.3% of fatal outcomes, whereas 97.7% of fatal outcomes reported in US clozapine-treated patients had another cause.

IMPLICATIONS/CONCLUSIONS

To help prevent future deaths in clozapine-treated patients, the clozapine package insert should focus on fatal outcomes during infections. Part II offers detailed solutions regarding current boxed warnings and lack of a warning for pneumonia and other infections. The Supplementary Material includes letters of support from 124 non-US clozapine experts from 44 countries/regions who support Parts I and II.